Spinal cord impairments, including spinal cord injury and spina bifida, have effects on a child's ability to ambulate. Spinal cord injuries in children often involve a traumatic injury to a region of the spinal cord, which results in impairments of motor, sensory, or both. Spina bifida is a congenital defect in which the neural tube does not close by the 28th day of gestation, affecting the thoracic, lumbar, or sacral region of the spinal cord. There are 2 classifications of spina bifida: occulta and aperta.1 In spina bifida occulta, the spinal cord and meninges are intact with no visible lesions and the vertebrae are not fused. Spina bifida aperta involves the protrusion of a cyst through the nonfused vertebrae and is separated into 2 subtypes: meningocele and myelomeningocele. In meningocele, the spinal cord is intact; however, the meninges and cerebrospinal fluid protrude through the vertebrae and are covered by skin. In myelomeningocele, the meninges, cerebrospinal fluid, and spinal cord are herniating through the nonfused vertebrae, are not covered by skin, and spinal nerve paralysis is common.
The level of injury to the spinal cord typically relates to functional capabilities and the amount of assistance the child requires for daily activities. The lower the lesion level, the more functional mobility the child is capable of achieving. Ambulation is often a primary goal for children with spinal cord impairments. Children who are able to ambulate are more independent in their activities of daily living and social roles.2
An increase in prenatal care and the introduction of folic acid has decreased the number of neural tube defects in the past 20 years; however, approximately 1500 children are born with spina bifida each year in the United States.3 In 2010, less than 10% of annual spinal cord injuries occurred in children younger than 15 years, with incidence rates rapidly increasing as age increased above 15 years.4 The most recent incidence rate reported for pediatric spinal cord injuries per year is 1.99 cases per 100 000 children in the United States.5.
There are no clinical practice guidelines focused on interventions to improve gait in children with spinal cord impairments. Current systematic reviews on pediatric interventions for ambulation have limited results for children with spinal cord injury and do not include spina bifida. There is one systematic review on locomotor training (LT) in pediatric spinal cord injury6; however, there are other intervention options to promote ambulation that warrant consideration. Scoping reviews have been described as a process of examining the existing literature to summarize research and identify research gaps. A scoping review typically investigates a broader topic than a systematic review and may or may not lead to a future systematic review.7 Therefore, the purpose of this scoping review was to provide a summary of the evidence for gait training in children with spinal cord impairments, specifically spinal cord injuries and spina bifida.
A scoping literature review was conducted to identify relevant citations from PubMed, Embase, and CINAHL.
The inclusion criteria were articles in English, human research, pediatric population (birth through the end of the 21st year), diagnosis of spinal cord impairment (spina bifida or spinal cord injury), and an intervention with gait as the outcome. The exclusion criteria included systematic reviews and scoping reviews.
Searches from PubMed, Embase, and CINAHL (Supplemental Digital Content 1, available at: http://links.lww.com/PPT/A182) were combined and screened and duplicates were removed (Figure). Two reviewers performed a title and abstract screen using predetermined inclusion and exclusion criteria, with an agreement of 87.1%. Two reviewers completed a full-text age screen with no discrepancies. Four reviewers then reviewed full texts using predetermined criteria. The articles were divided into 2 groups, and 2 reviewers independently examined each group. Agreement among 4 reviewers was 83.3%. A third reviewer decided any discrepancies. Each reviewer completed full-text references searches, which were confirmed by a second reviewer.
Four reviewers were each assigned an intervention category and extracted data from the included articles. Each reviewer extracted the following information—author, year of publication, subjects (age range and diagnosis), intervention, results, and level of evidence—to be recorded in a summary table. Only the results for gait outcomes were extracted for this review including level of ambulation, cadence, velocity, step length, stride length, speed, step rate, 6-minute walk test, distance, energy consumption, gait efficiency, endurance, total number of steps, leg activity, and interlimb stepping patterns. Quantitative data were unable to be extracted from the articles for comparison because each study used different outcome measures; therefore, reliability and percent agreements were not computed.
Level of Evidence
The level of evidence for each article was determined using the Oxford Center for Evidence-Based Medicine Levels of Evidence 2 Table.8 This tool was developed by leaders in the field of evidence-based medicine to formulate a standardized classification that could be used in a clinical setting to determine recommendation grades from level of evidence.9 Quasiexperimental studies and case studies, which were not included in the Oxford Center for Evidence-Based Medicine, were classified as level III and level V evidence, respectively, for the purpose of this review.10
The initial search of the databases yielded 697 articles, and after duplicates were removed, 580 articles remained. The title and abstract screen resulted in 73 included articles. A full-text age screen left 48 articles to be included. After the full-text review, 13 articles remained. Hand searches yielded 13 articles, which brought the total number to 26 articles (Figure). There were 271 children (age ≤ 21) within the 26 articles, which consisted primarily of studies conducted between the years of 1981 and 2013, with spinal cord impairments that received one of the following interventions for gait: orthotics, use of an assistive device, electrical stimulation, treadmill training, and infant treadmill training.
According to the Oxford Levels of Evidence 2 Table,8 only 1 article was a randomized control trial and classified as level II. The majority of articles (77%) were level III and IV evidence (13 and 7, respectively). Five articles were level V evidence.
Ten articles discussed the use of orthoses or assistive devices for ambulation in children with spinal cord impairments (Table). A variety of orthotic/assistive devices were studied, including ankle-foot orthoses (AFOs), hip-knee-ankle-foot orthoses (HKAFOs), reciprocating gait orthoses (RGOs), parapodiums, swivel walkers, hip guidance orthoses (HGO), and forearm crutches. Walkers and standard crutches were discussed when used in association with orthoses but were not specifically studied. Solid AFOs were recommended for children with lower lumbar and sacral level lesions, although concern for AFOs creating abnormal knee stresses was noted.11 Forearm crutches in addition to AFOs were recommended when more support was necessary, especially for longer distances. Seven studies investigated more supportive orthoses for children with thoracic or lumbar level lesions. These had a greater variety in orthotic type, and the literature to support each was inconsistent. Compared with a parapodium, a swivel walker allowed freedom of the patients' upper extremities, less energy consumption, enhanced distance efficiency, and was preferred by most patients and families; however, children demonstrated a slower walking velocity.12 There was conflicting evidence about whether RGOs or HKAFOs promoted a faster, more energy-efficient gait.13,14 Long-term studies demonstrated improvements in ambulatory status with either RGO or HKAFO use.15–17 In long-term studies, approximately half of the participants began using a wheelchair for long-distance mobility at the time of follow-up; however, this should not imply failure, as orthoses should still be used to encourage any amount of ambulation because of the many physical and social benefits.15–17
In these 10 studies, a maximum of 2 devices were compared in each study; therefore, one orthosis cannot be recommended as superior for this population. Therefore, choosing an appropriate orthotic device should be an individualized decision for each child. The selection of a device should not be limited to gait and energy parameters, but also consideration of ease of donning/doffing, transfers, ability to make modifications for scoliosis or hip flexion contractures, the need for an assistive device, the child's environment, and musculoskeletal management. To increase compliance, it is also important to incorporate patient and family motivation and satisfaction with the orthotic device. Children with spinal cord impairments may not use ambulation as their primary means of mobility; however, it is still recommended for children to use orthoses to ambulate short distances.
Seven articles discussed electrical stimulation as an intervention for children with spinal cord impairments, and 5 articles examined the use of functional electrical stimulation (FES), via percutaneous or implantable systems. Three articles compared FES to bracing (Table). All of the FES studies were performed on subjects with spinal cord injury. For the implantable FES systems, a considerable period of immobilization was required postsurgery to prevent excessive joint movement to maintain electrode placement. Following immobilization, FES training was provided.18,19 Both intramuscular and percutaneous systems required extensive conditioning and training that ranged from 5 to 12 weeks. Only Bonaroti et al20 and Johnston et al21 examined gait as a primary outcome, whereas other studies focused on upright mobility tasks. Therefore, more research is required with gait as a primary outcome. All subjects in these studies also wore AFOs or a more restrictive brace. Overall, the studies found that FES may increase gait velocity and distance when used in conjunction with orthotics; however, it may not change the level of independence in ambulation achieved. The length of time required for FES training and the invasive nature of some of the interventions must be considered, as well as the cost and lack of research on the long-term physiologic effects of the system when deciding whether this is an appropriate intervention.
Two articles examined electrical stimulation in the spina bifida population. One article on an 8-week neuromuscular electrical stimulation program used gait as a functional measure, but it was not the primary outcome.22 The second article on threshold nighttime stimulation examined gait through observational review; however, because of the intensity of the program only 7 of 15 subjects completed 9 months, and none completed a year.23 Both articles demonstrated an increase in walking velocity; however, more research must be conducted examining program intensity with a primary focus on gait, using standardized measures, to determine the benefit of these interventions in improving gait in children with spina bifida.
Six articles discussed the use of treadmill training for ambulation in children with spinal cord impairments (Table). An LT program led to improvements in speed, distance, use of less restrictive assistive devices, capacity for walking (number of steps per hour), and community-based activity (number of steps per day). These improvements were demonstrated in individuals with spinal cord injuries at various functional baselines.24–26 Locomotor training protocols, including body weight support, overground training, and community reintegration, used a large number of sessions to maximize the effects of neuroplasticity. The gains achieved with treadmill training could be maintained, and one individual continued to make gains long term.27 Implementing an LT program early in inpatient rehabilitation required modifications to the protocol, such as adding upper extremity support.28 When these modifications were removed, more facilitation was necessary, but the modified program was beneficial because it allowed earlier implementation of LT. In the children with spina bifida, a treadmill training program following a 12-step protocol of increased work and duration was shown to be an effective intervention to increase walking speed, total distance, and endurance.29 Treadmill training is an effective intervention because it teaches a functional skill that is used daily by the individual. It also increases speed, distance, and community ambulation, in addition to, the use of less restrictive assistive devices. Treadmill training is recommended as an intervention for functional gains.
Infant Treadmill Training
Three articles discussed facilitation of infant stepping on a treadmill (Table). The articles focused on infants with myelomeningocele who had lumbar or sacral level lesions. All articles followed a similar treadmill program with infants supported in the upright position while stepping. Step rate or total number of steps, leg activity, and interlimb stepping patterns were measured throughout the infant's first year. Pantall et al30 and Moerchen and Hoefakker31 both extended the research of Teulier et al32 and examined the response of infant stepping with enhanced sensory input.
Treadmill training alone did not increase step rate for infants with myelomeningocele; however, it increased leg activity on a moving treadmill.32 Increased step rate was observed with the addition of sensory enhancements to treadmill training, specifically friction, visual flow, combining friction, and loading, as well as manual assistance.30,31 Early intervention for walking that incorporates task-specific sensory input and promotes activity-dependent neuroplasticity may aid in decreasing the delay in walking that children with myelomeningocele often experience. However, no long-term prospective studies have been conducted that examine the effect of infant treadmill training on walking velocity or initiation of walking. Future research on this topic is necessary to determine whether infant treadmill training decreases delays in the onset of walking and increases the capability to maintain ambulation further into adulthood. No negative consequences were found with infant treadmill training; therefore, if accessibility and time permit, it can be an effective intervention for increasing leg activity, which may lead to increased strength and development of graded control for ambulation.
The purpose of this scoping review was to provide a summary of the literature and synthesize the research on interventions used to improve gait in children with spinal cord impairments, to direct clinical practice and highlight areas for future research. In this review, 4 broad categories of interventions were identified: orthoses/assistive devices, electrical stimulation, treadmill training, and infant treadmill training. These interventions have variable success in children with spinal cord impairments, and subsequently, the choice of intervention must be individualized to the children on the basis of their unique physical, environmental, and social needs, as well as, considering the facility's characteristics and resources (Table).
The majority of research was level III and level IV evidence; therefore, the conclusions may not be as strongly supported as they could be with future, higher-quality research. In this review, outcomes were included if they were specific to gait, including, but not limited to, level of ambulation, cadence, velocity, step length, and 6-minute walk test. Studies that measured only muscle activity, oxygen consumption, energy cost, and kinetic or kinematic factors from 3D gait analysis were excluded because each factor's effect on gait was not directly shown in the data. Therefore, expanding the search criteria to include these factors may yield additional beneficial information that could indirectly influence gait outcomes.
The 26 studies reported small sample sizes and often had large degrees of variability among participants. The largest sample size was 41 participants; however, the majority of studies had considerably fewer participants, including 5 case studies (Table). Smaller sample sizes were expected within a pediatric population because of the difficulty with recruiting children, obtaining parental consent, and availability for follow-up.23 These smaller sample sizes and loss of follow-up data make it difficult to generalize results. More research is required to draw more definitive conclusions.
Articles included in this study suggest the interventions led to improvements in gait; however, these effects could also have been influenced by other mediating factors. Typically, children with lower level lesions have a better prognosis for independent community ambulation.13 The natural progression of recovery may be a contributing factor to the higher return of function seen in children with lower level injuries. Furthermore, as children increase in age and size, their ambulation may decline or require the need for increased external support15,16 because of changing biomechanics. In addition, the natural gross motor developmental sequence may have affected the results of the studies on infant treadmill training. As infants were learning more meaningful motor activities, such as stepping, the researchers observed a decrease in non-purposeful leg activity.30
Synthesizing the results of the reviewed studies, with consideration of each intervention's clinical implications, led to the recommendation of using an appropriate orthotic support for the child in combination with treadmill training and/or electrical stimulation. Orthoses are recommended for clinical use, with consideration of level of injury, the amount of support required, appropriate fit, and patient compliance. Although many studies have shown that orthoses can increase velocity and decrease energy expenditure, the children still functioned below average compared with age-matched peers. As a child grows, transitioning to a different orthotic device, or including the use of a wheelchair in some environments, may be beneficial to allow the child to keep pace with peers and function independently in his or her environment.15 It is important to review what devices are currently available and used clinically, and allow this to guide future research, including ongoing advances in orthotic devices, because the existing research on orthotics consists of primarily studies conducted between the years of 1981 and 2001. Specifically examining the use of walkers or other assistive devices may be beneficial to determine whether they offer more support or earlier attainment of walking when used independently or in conjunction with orthoses. Although FES demonstrated increases in gait velocity and distance ambulated and was preferable to long leg braces for cosmesis and donning/doffing, the high cost and the procedure for receiving electrical stimulation may not be conducive to immediate use in all clinics (Table). Therefore, with potentially wider availability of equipment, treadmill training may be more easily initiated if equipment is available. Treadmill training is recommended for clinical use because it leads to improvements in gait, including increases in speed, distance, or the level of community reintegration. Infant treadmill stepping in the spina bifida population showed promising effects in increasing leg activity; however, long-term effects have not yet been studied. It is important to consider individualizing treatment for each child on the basis of lesion level, current and previous level of function, and ambulation goals. It is essential to consider support of optimal musculoskeletal alignment and integrity over time, as well as physical, social, and environmental factors.
A limitation of the search strategy was that 13 of 26 articles were identified through hand searches. Although an exhaustive search was attempted, there may be additional articles that fit the criteria that were not included. The limitations of data extraction included 4 reviewers examining the studies and identifying appropriate outcomes for gait interventions. It was not possible to prespecify gait-related outcomes to report because of the wide variety of reported outcomes. Levels of evidence were determined by one reviewer for their respective intervention category, and may have variability because of the lack of using a second reviewer. The Oxford Levels of Evidence Table 18 was not sufficiently inclusive for the types of studies identified in this review; therefore, the quasiexperimental and case study subgroups had to be added.
This scoping review offers a contribution to current research because a study specific to gait training for the population of pediatric spinal cord impairment has not previously been reported. Although this review was able to offer suggestions for the use of select interventions for this highly diverse population, more research is required to determine definitive clinical practice guidelines. Areas of future research that can build on current literature and may be beneficial include long-term implantable FES for varying lesion levels, effective orthoses currently used for children with high-level lesions, new advances in orthoses, use of walkers or other assistive devices, low-intensity outpatient treadmill training, and prospective longitudinal studies in the area of infant treadmill training. In future research, it will be helpful to use standardized measures of gait throughout all studies, such as ambulation velocity, step rate, energy consumption, and distance ambulated, to allow comparison across interventions as well as quantitative data analysis, which can help inform a future systematic review.
We would like to thank Chad Cook, PT, PhD, and Leila Ledbetter, MLIS, for their assistance with this project.
1. Campbell SK, Palisano RJ, Orlin M. Physical Therapy for Children. 4th ed. St Louis, MO: Saunders; 2012.
2. Zwicker JG, Mayson TA. Effectiveness of treadmill training in children with motor impairments: an overview of systematic reviews. Pediatr Phys Ther. 2010;22(4):361–377.
3. Parker SE, Mai CT, Canfield MA, et al Updated National Birth Prevalence estimates for selected birth defects in the United States, 2004-2006. Birth Defects Res A Clin Mol Teratol. 2010;88(12):1008–1016.
4. Parent S, Mac-Thiong JM, Roy-Beaudry M, Sosa JF, Labelle H. Spinal cord injury
in the pediatric population: a systematic review of the literature. J Neurotrauma. 2011;28(8):1515–1524.
5. Vitale MG, Goss JM, Matsumoto H, Roye DP Jr. Epidemiology of pediatric spinal cord injury
in the United States: years 1997 and 2000. J Pediatr Orthop. 2006;26(6):745–749.
6. Gorski K, Harbold K, Haverstick K, Schultz E, Shealy SE, Krisa L. Locomotor training in the pediatric spinal cord injury
population: a systematic review of the literature. Top Spinal Cord Inj Rehabil. 2016;22(2):135–148.
7. Arksey H, O'Malley L. Scoping studies: towards a methodological framework. Int J Soc Res Methodol. 2005;8(1):19–32.
8. Centre for Evidence-based Medicine. Levels of Evidence (May 2001). Oxford, England: Centre for Evidence-based Medicine; 2009.
9. Grondin SC, Schieman C. Evidence-based medicine: levels of evidence and evaluation systems. In: Ferguson KM, ed. Difficult Decisions in Thoracic Surgery: An Evidence-Based Approach. London, England: Springer; 2011:13–22.
10. Sackett D, Straus SE, Richardson WS, et al Evidence-Based Medicine: How to Practice and Teach EBM. Toronto, Canada: Churchill Livingstone; 2000.
11. Thomson JD, Ounpuu S, Davis RB, DeLuca PA. The effects of ankle-foot orthoses on the ankle and knee in persons with myelomeningocele: an evaluation using three-dimensional gait
analysis. J Pediatr Orthop. 1999;19(1):27–33.
12. Lough LK, Nielsen DH. Ambulation of children with myelomeningocele: parapodium versus parapodium with Orlau swivel modification. Dev Med Child Neurol. 1986;28(4):489–497.
13. Katz DE, Haideri N, Song K, Wyrick P. Comparative study of conventional hip-knee-ankle-foot orthoses versus reciprocating-gait
orthoses for children with high-level paraparesis. J Pediatr Orthop. 1997;17(3):377–386.
14. Cuddeford TJ, Freeling RP, Thomas SS, et al Energy consumption in children with myelomeningocele: a comparison between reciprocating gait
orthosis and hip-knee-ankle-foot orthosis ambulators. Dev Med Child Neurol. 1997;39(4):239–242.
15. Thomas SS, Buckon CE, Melchionni J, Magnusson M, Aiona MD. Longitudinal assessment of oxygen cost and velocity in children with myelomeningocele: comparison of the hip-knee-ankle-foot orthosis and the reciprocating gait
orthosis. J Pediatr Orthop. 2001;21(6):798–803.
16. Gerritsma-Bleeker CLE, Heeg M, Vos-Niel H. Ambulation with the reciprocating-gait
orthosis: experience in 15 children with myelomeningocele or paraplegia. Acta Orthop Scand. 1997;68(5):470–474.
17. Guidera KJ, Smith S, Raney E, et al Use of the reciprocating gait
orthosis in myelodysplasia. J Pediatr Orthop. 1993;13(3):341–348.
18. Johnston TE, Betz RR, Smith BT, Mulcahey MJ. Implanted functional electrical stimulation: an alternative for standing and walking in pediatric spinal cord injury
. Spinal Cord. 2003;41(3):144–152.
19. Betz RR, Johnston TE, Smith BT, Mulcahey MJ, McCarthy JJ. Three-year follow-up of an implanted functional electrical stimulation system for upright mobility in a child with a thoracic level spinal cord injury
. J Spinal Cord Med. 2002;25(4):345–350.
20. Bonaroti D, Akers J, Smith BT, Mulcahey MJ, Betz RR. A comparison of FES with KAFO for providing ambulation and upright mobility in a child with a complete thoracic spinal cord injury
. J Spinal Cord Med. 1999;22(3):159–166.
21. Johnston TE, Finson RL, Smith BT, Bonaroti DM, Betz RR, Mulcahey MJ. Technical perspective. Functional electrical stimulation for augmented walking in adolescents with incomplete spinal cord injury
. J Spinal Cord Med. 2003;26(4):390–400.
22. Karmel-Ross K, Cooperman DR, Van Doren CL. The effect of electrical stimulation on quadriceps femoris muscle torque in children with spina bifida
. Phys Ther. 1992;72(10):723–730.
23. Walker JL, Ryan SW, Coburn TR. Does threshold nighttime electrical stimulation benefit children with spina bifida
? A pilot study. Clin Orthop Relat Res. 2011;469(5):1297–1301.
24. Behrman AL, Watson E, Fried G, et al Restorative rehabilitation entails a paradigm shift in pediatric incomplete spinal cord injury
in adolescence: an illustrative case series. J Pediatr Rehabil Med. 2012;5(4):245–259.
25. O'Donnell CM, Harvey AR. An outpatient low-intensity locomotor training programme for paediatric chronic incomplete spinal cord injury
. Spinal Cord. 2013;51(8):650–651.
26. Behrman AL, Nair PM, Bowden MG, et al Locomotor training restores walking in a nonambulatory child with chronic, severe, incomplete cervical spinal cord injury
. Phys Ther. 2008;88(5):580–590.
27. Fox EJ, Tester NJ, Phadke CP, et al Ongoing walking recovery 2 years after locomotor training in a child with severe incomplete spinal cord injury
. Phys Ther. 2010;90(5):793–802.
28. Prosser LA. Locomotor training within an inpatient rehabilitation program after pediatric incomplete spinal cord injury
. Phys Ther. 2007;87(9):1224–1232.
29. de Groot JF, Takken T, van Brussel M, et al Randomized controlled study of home-based treadmill training for ambulatory children with spina bifida
. Neurorehabil Neural Repair. 2011;25(7):597–606.
30. Pantall A, Teulier C, Smith BA, Moerchen V, Ulrich BD. Impact of enhanced sensory input on treadmill step frequency: infants born with myelomeningocele. Pediatr Phys Ther. 2011;23(1):42–52.
31. Moerchen VA, Hoefakker HL. Infants with spina bifida
: immediate responses to contextual and manual sensory augmentation during treadmill stepping. Pediatr Phys Ther. 2013;25(1):36–45.
32. Teulier C, Smith BA, Kubo M, et al Stepping responses of infants with myelomeningocele when supported on a motorized treadmill. Phys Ther. 2009;89(1):60–72.
33. Rose GK, Stallard J, Sankarankutty M. Clinical evaluation of spina bifida
patients using hip guidance orthosis. Dev Med Child Neurol. 1981 Feb;23(1):30–40.
34. Vankoski S, Moore C, Statler KD, Sarwark JF, Dias L. The influence of forearm crutches on pelvic and hip kinematics in children with myelomeningocele: Don't throw away the crutches. Dev Med Child Neurol. Sep 1997;39(9):614–619.
35. Duffy CM, Graham HK, Cosgrove AP. The influence of ankle-foot orthoses on gait
and energy expenditure in spina bifida
. J Pediatr Orthop. May-Jun 2000;20(3):356–361.
36. Bonaroti D, Akers JM, Smith BT, Mulcahey MJ, Betz RR. Comparison of functional electrical stimulation to long leg braces for upright mobility for children with complete thoracic level spinal injuries. Arch Phys Med Rehabil. 1999;80(9):1047–1053.